A migration utilizing ONware™ frees your application and data from the limitations of database, physical data structure, hardware, operating system, and user interface, facilitating access to your data and integration of your application with the tools and solutions of your choice. At the conclusion of this process your application:

• Is enabled to run on any type of database, independent of the environment in which it was originally developed.
• Can run on a network of computers with differing operating system types instead of being limited to operating on a single computer with a single operating system.
• Is independent of the hardware, operating system, database, and user interface.
• Can perform I/O with multiple database types, concurrently, located on any node on the network, without re-engineering the original application.
• May be deployed in all combinations of multi-tier architecture (host centric, thin client, fat client, distributed).
• Integrates with tools and solutions that are supported by the databases of your choice such as .Net, JAVA, Web tools, Business Intelligence tools, and more.

Benefits include:

• Eliminate dependency on any one database, operating system or hardware environment.
• One set of application code operates directly on multiple platforms, with many databases, without change.
• Developers continue to work in a familiar environment while also learning new relational tools.
• Choice of development tools and package solutions opens up to the world of relational offerings.
• Facilitate access to data by making it available in standard relational formats.
• Corporate use of diverse database and operating systems may be consolidated.
• Distribute your applications across any configuration of a multi-tiered architecture.

Application Redeployment on ONware

Application Redeployment is the migration of the MultiValue System to the ONware technology independent platform. ONgroup’s methodology involves isolating the application with appropriate test data and test suites. Once isolated the application is imported, compiled, and tested as required to have the application fully operational. Data is replicated as relational tables.

At the completion of this phase, all data resides within the Relational Database Management System (RDBMS) tables of your choice. The MultiValue application, programs, dictionaries, PROCs, paragraphs and sentences, are operational regardless of what database now stores the data. The applications will still perform I/O with your MultiValue database if you see value in prolonging the life of the MultiValue platform.

The flexible structure provided by ONware allows the migration of data one file at a time as needs dictate. For example, convert the active portion of your database to a Relational database and then migrate the older transaction history over an extended period. This facilitates the migration of very large amounts of historical data.

Data Reorganization (Normalization)

Some relational database tools require a normalized data structure. The challenge with applications originally written with a MultiValue environment is that they understand a data structure that differs from standard relational data structures. ONware solves this challenge for you. ONware understands the data structure that the MultiValue application expects. ONware allows the data be stored physically in standard relational data structures even though the applications expect the data to be in a MultiValue format.

The benefit is that ONware manages the difference in data structures on the fly, allowing the application to perform as it did in the original MultiValue environment even though the physical structure of the data is a normalized format to comply with standard relational conventions.

ONware technology allows:

• Nested MultiValue and even sub-multivalue data structures to be projected into first normal form
• Repositioning of data fields and conversion of data with internal coding as required. This allows integration of multiple applications with differing views of the database.

Many organizations we work with find that when they review how the data is to be used, only certain files require normalization. Normalization of a table may occur at anytime in the future. It may, therefore, be more efficient to normalize files as requirements dictate.

Other organizations may want to normalize the complete database as one project. The best approach will depend on your business needs and on the state of your data. ONware includes tools to automate the normalization process and is available to assist with the initial analysis of file structures and the integrity of the data. Data typing is typically more stringent with relational databases than it is with MultiValue database. Some data cleansing may, therefore, be required to ensure that all data fits data integrity constraints.